Readmissions Among Patients With Severe Sepsis/Septic Shock Among Inner-City Minority New Yorkers

Abstract

PURPOSE: Readmissions and health care utilization among patients admitted with severe sepsis/ septic shock (SS) is not well described in literature compared to other conditions like CHF and pneumonia. Readmissions are a particularly significant problem in inner-city areas. We report the readmissions and outcomes of SS patients in a Medical-ICU(MICU) in an inner-city hospital that serves inner-city minority New-Yorkers. METHODS: A retrospective analyses of 127 patients admitted to the MICU in 2011 with a primary diagnosis of Severe Sepsis/Septic Shock was performed. Data collected included demographics, clinical, ICU and hospital length of stay, readmissions (30, 90 and 180 days) and deaths. Logistic regression analysis was used to analyze and p-values are reported. RESULTS: Of the 127 patients, 61(48%) were females and 66(52%) males. The median age of the patient was 58years, median BMI was 26kg/m2 and median APACHE II score was 22(range 6-42). 49 patients (39%) had Diabetes; 47(37%) had Hypertension; 24(19%) had HIV-infection, and 4(3%) has ESRD. 92(72%) required mechanical-ventilation, and the median number of days on ventilator in this group was 4(range 0-35) days. The Median-LOS in ICU was 6(range1-25) days and in-hospital was 10(range 1-114) days. 32patients(25%) died in-hospital. Of the 95patients who survived the index admission, 25(26%) experienced at least one readmission within 30days; 36(38%) were readmitted within 90 days; and 41(46%) within 6 months after index admission.In multivariate analysis, older age(OR1.03,p-value=0.13) and higher APACHE-II scores(OR1.12;p-value=0.00) were significantly associated with in-hospital mortality. Age, gender, BMI, APACHE-II scores, mechanical-ventilator days, LOS in index-admission did not predict readmission at 30,90 and 180 days(p=NS). CONCLUSIONS: This is the largest reported series on re-admissions among patients with severe-sepsis/septic shock. Although significant high hospital readmission rates were noted at 30, 90 and 180 days, it was not associated with any demographic /clinical variables in this study.